JPWO2018220787A1 - Map processing apparatus, map processing method, and map processing program - Google Patents

Abstract

The primary vector calculation unit (23) includes, for each of the first map and the second map indicating the existence probability that an object exists in each region, the existence probability for the target region and the existence probability for the adjacent region adjacent to the target region. And the sum of the vectors for the surrounding regions of the target region is calculated as the primary vector of the target region. The secondary vector calculation unit (24) calculates the sum of the primary vectors of the respective areas included in the attention area as the secondary vector of the attention area. The determination unit (25) compares the secondary vector calculated for the attention area for the first map with the secondary vector calculated for the attention area for the second map, and determines the first map. It is determined whether or not the attention area for and the attention area for the second map correspond to each other.

Description

  The present invention relates to a technique for associating a plurality of maps indicating the probability of existence of an object in each region.

  In order to obtain a map of a wide range around a moving object such as a vehicle, a map acquired by the moving object may be combined with a map acquired by a moving object or a roadside device existing in the vicinity. Patent Documents 1 and 2 describe a method of synthesizing a map.

  In Patent Document 1, for each map, each grid is divided into an occupied grid, an unoccupied grid, and an unknown grid. A histogram representing the number of occupied grids, unoccupied grids, and unknown grids is created for a window centered on the occupied grid. Based on the histogram, corresponding points between the maps are identified. Then, coordinate conversion is performed so that the specified points overlap each other.

  In Patent Document 2, the distance between obstacle cells is calculated. And the optimization process regarding the total value of the calculated distance is performed, and coordinate conversion is performed.

JP 2005-326944 A JP 2009-157430 A

In the method described in Patent Literature 1, since processing based on a histogram is performed, there are many calculation processes and processing time is required. Further, in the method described in Patent Document 2, when an obstacle moves, noise increases and the synthesis accuracy is lowered.
An object of the present invention is to make it possible to specify corresponding points between maps with high accuracy even when a moving object exists while shortening the processing time.

The map processing apparatus according to this invention is
For each of the first map and the second map indicating the existence probability that an object exists in each area, the existence probability for the target area and the adjacent area adjacent to the target area with at least a part of the area as the target area A first vector calculation unit that calculates a difference between the existence probability of the target region as a vector for the adjacent region with respect to the target region and a sum of vectors with respect to surrounding regions for the target region as a primary vector of the target region When,
For each of the first map and the second map, two or more regions are used as a region of interest, and the sum of the primary vectors of the regions included in the region of interest is calculated as a secondary vector of the region of interest 2 A next vector calculation unit;
Comparing the secondary vector calculated for the region of interest for the first map with the secondary vector calculated for the region of interest for the second map; A determination unit that determines whether or not the region of interest for and the region of interest for the second map correspond to each other.

  In the present invention, the corresponding points between the maps are specified by comparing the vectors by using the difference in the existence probability that the object exists in each region as a vector. Thereby, it is possible to specify the corresponding points between maps with high accuracy even when there is a moving object while shortening the processing time.

1 is a configuration diagram of a map processing apparatus 10 according to Embodiment 1. FIG. 2 is a flowchart of overall processing of the map processing apparatus 10 according to the first embodiment. Explanatory drawing of the 1st map 31 and the 2nd map 32 which concern on Embodiment 1. FIG. FIG. 3 is an explanatory diagram of resolution change processing according to the first embodiment. 4 is a flowchart of primary vector calculation processing according to the first embodiment. Explanatory drawing of the object area | region selection process which concerns on Embodiment 1. FIG. Explanatory drawing of the vector 41 which concerns on Embodiment 1. FIG. Explanatory drawing of the primary vector 42 concerning Embodiment 1. FIG. 4 is a flowchart of secondary vector calculation processing according to the first embodiment. Explanatory drawing of the 1st attention area 38 which concerns on Embodiment 1. FIG. Explanatory drawing of the secondary vector 43 which concerns on Embodiment 1. FIG. 6 is a flowchart of similar region search processing according to the first embodiment. Explanatory drawing of the 2nd attention area | region 39 which concerns on Embodiment 1. FIG. Explanatory drawing of 1st attention area | region 38 'adjacent to the 1st attention area | region 38 which concerns on Embodiment 1. FIG. Explanatory drawing of 1st attention area | region 38 'adjacent to the 1st attention area | region 38 which concerns on Embodiment 1. FIG. Explanatory drawing of 1st attention area | region 38 'adjacent to the 1st attention area | region 38 which concerns on Embodiment 1. FIG. Explanatory drawing of the process which rotates the 1st map 31 which concerns on Embodiment 1. FIG. The block diagram of the map processing apparatus 10 which concerns on the modification 3. FIG. The block diagram of the map processing apparatus 10 which concerns on Embodiment 2. FIG. 10 is a flowchart of overall processing of the map processing apparatus 10 according to the second embodiment. The block diagram of the map processing apparatus 10 which concerns on Embodiment 3. FIG. 10 is a flowchart of overall processing of the map processing apparatus 10 according to the third embodiment.

Embodiment 1 FIG.
*** Explanation of configuration ***
With reference to FIG. 1, the structure of the map processing apparatus 10 which concerns on Embodiment 1 is demonstrated.
The map processing device 10 is a computer.
The map processing apparatus 10 includes hardware including a processor 11, a memory 12, a storage 13, and a communication interface 14. The processor 11 is connected to other hardware via a signal line, and controls these other hardware.

  The processor 11 is an IC (Integrated Circuit) that performs processing. The processor 11 is a CPU (Central Processing Unit), a DSP (Digital Signal Processor), or a GPU (Graphics Processing Unit) as specific examples.

  The memory 12 is a storage device that temporarily stores data. As a specific example, the memory 12 is an SRAM (Static Random Access Memory) or a DRAM (Dynamic Random Access Memory).

  The storage 13 is a storage device that stores data. The storage 13 is an HDD (Hard Disk Drive) as a specific example. The storage 13 is a portable storage such as an SD (Registered Trademark, Secure Digital) memory card, CF (CompactFlash), NAND flash, flexible disk, optical disk, compact disk, Blu-ray (registered trademark) disk, or DVD (Digital Versatile Disk). It may be a medium.

  The communication interface 14 is an interface for communicating with an external device. As a specific example, the communication interface 14 is a port of Ethernet (registered trademark), USB (Universal Serial Bus), or HDMI (registered trademark, High-Definition Multimedia Interface).

The map processing apparatus 10 includes an acquisition unit 21, a resolution change unit 22, a primary vector calculation unit 23, a secondary vector calculation unit 24, and a determination unit 25 as functional components. The function of each functional component of the map processing apparatus 10 is realized by software.
The storage 13 stores a program that realizes the function of each functional component of the map processing device 10. This program is read into the memory 12 by the processor 11 and executed by the processor 11. Thereby, the function of each function component of the map processing apparatus 10 is implement | achieved.

  In FIG. 1, only one processor 11 is shown. However, the map processing apparatus 10 may include a plurality of processors that replace the processor 11. The plurality of processors share the execution of a program that realizes the function of each functional component of the map processing device 10. Each processor is an IC that performs processing in the same manner as the processor 11.

*** Explanation of operation ***
The operation of the map processing apparatus 10 according to the first embodiment will be described with reference to FIGS.
The operation of the map processing apparatus 10 according to the first embodiment corresponds to the map processing method according to the first embodiment. The operation of the map processing apparatus 10 according to the first embodiment corresponds to the processing of the map processing program according to the first embodiment.

With reference to FIG. 2, the overall processing of the map processing apparatus 10 according to the first embodiment will be described.
(Step S10: Acquisition process)
The acquisition unit 21 acquires the first map 31 and the second map 32 as the maps to be combined.
Specifically, the acquisition unit 21 acquires the first map 31 and the second map 32 from an external device via the communication interface 14. Alternatively, the acquisition unit 21 acquires the first map 31 and the second map 32 stored in the memory 12 or the storage 13 in advance.

With reference to FIG. 3, the 1st map 31 and the 2nd map 32 which concern on Embodiment 1 are demonstrated.
The first map 31 and the second map 32 are maps showing the existence probability that an object exists in each area 33.
In the first embodiment, as shown in FIG. 3, the first map 31 and the second map 32 are divided into a plurality of regions 33 in a lattice area within the map range, and the existence probability that an object exists in each region 33 is high. It is the occupancy grid map shown. In the first embodiment, the existence probability of each area is “1” (occupation) indicating that an object exists, “0” (vacant) indicating that no object exists, and whether or not an object exists. It is either “0.5” (unknown) indicating unknown. In FIG. 3, the region having the existence probability “1” is indicated by rhombus hatching, the region having the existence probability “0” is indicated by white, and the region having the existence probability “0.5” is indicated by hatching. .

The first map 31 is a map generated by a moving body such as a vehicle, for example. Moreover, the 2nd map 32 is a map produced | generated by the peripheral body which is another mobile body etc. which are different from the mobile body which produced | generated the 1st map 31, for example.
Specifically, the moving body acquires point cloud data around the moving body using a sensor such as a stereo camera or a laser sensor. Then, the moving object calculates the existence probability that the object exists in each region 33 obtained by dividing the periphery of the moving object into a lattice from the acquired point cloud data. The first map 31 is generated by repeating this process while the moving body is moving. Similarly, the second map 32 is generated by repeatedly performing the process of acquiring the point cloud data and calculating the existence probability that the object exists at each position while the peripheral body moves.

  In Embodiment 1, it is assumed that the positions of the regions 33 of the first map 31 and the second map 32 are specified. As described above, when the first map 31 and the second map 32 are generated by the moving body, the first map is determined based on the position of the moving body specified by the positioning device mounted on the moving body and the sensor information. The position of each area | region 33 of 31 and the 2nd map 32 is pinpointed. In the first map 31 and the second map 32, the position of each region 33 is represented in the global coordinate system.

(Step S20: Resolution changing process)
The resolution changing unit 22 reduces the resolution of the first map 31 and the second map 32 by making a plurality of areas into one area for the first map 31 and the second map 32.

This will be specifically described with reference to FIG. FIG. 4 shows an example in which the resolution of the first map 31 is reduced. The second map 32 is also reduced in resolution by the same method.
The resolution changing unit 22 divides each area 33 of the first map 31 and the second map 32 into new areas 34 for each designated magnification range from the reference position. In FIG. 4, a total of four regions 33, two vertically and two horizontally, are defined as one new region 34. The resolution changing unit 22 determines the existence probability for each new area 34 as follows. (1) The resolution changing unit 22 determines the existence probability to be “1” when there is at least one region 33 having the existence probability “1”. (2) The resolution changing unit 22 determines the existence probability to be “0” when the existence probability of all the regions 33 is “0”. (3) The resolution changing unit 22 determines the existence probability to be “0.5” when there is at least one region 33 having the existence probability of “0.5”.

(Step S30: primary vector calculation process)
The primary vector calculation unit 23 calculates a primary vector 42 for the first map 31 whose resolution has been reduced in step S20.

With reference to FIG. 5, the primary vector calculation process according to the first embodiment will be described.
(Step S301: Determination area selection process)
As illustrated in FIG. 6, the primary vector calculation unit 23 selects at least a part of the region 34 as the determination region 35 from the first map 31 whose resolution has been reduced in step S <b> 20.
The position of each area 33 of the first map 31 and the second map 32 is specified. Therefore, the position of each region 34 is also specified. Therefore, the primary vector calculation unit 23 can generally specify which part of the first map 31 and which part of the second map 32 overlap from the position of the region 34. Therefore, the primary vector calculation unit 23 selects a partial region 34 of the first map 31 that is likely to overlap the second map 32 as the determination region 35.
Here, the primary vector calculation unit 23 selects a target number of regions 34 from the outside as a determination region 35 for one side of the rectangular first map 31. In FIG. 6, three areas 34 from the outside are selected as determination areas 35 for the left side. The outermost region 34 is excluded from the determination region 35 because a primary vector 42 described later cannot be calculated. The number of objects is determined by the accuracy of the position of the region 34, for example.

(Step S302: Object extraction process)
The primary vector calculation unit 23 extracts a region 34 among the regions 34 selected as the determination region 35 in step S <b> 301 as the target region 36.

数１では、ベクトルａ^→ _０は対象領域３６の存在確率である。ベクトルａ^→ _ｉｊは隣接領域３７の存在確率である。変数ｉは横方向の領域３４の位置を表し、変数ｊは縦方向の領域３４の位置を表す。したがって、ベクトル４１は、（ａ^→ _ｉｊ−ａ^→ _０）である。ベクトルｂ^→ _０は対象領域３６の１次ベクトル４２である。(Step S303: Vector calculation process)
The primary vector calculation unit 23 sets the difference between the existence probability for the target region 36 and the existence probability for the adjacent region 37 in the region 34 adjacent to the target region 36 as a vector 41 for the adjacent region 37 for the target region 36. calculate. As a specific example, as shown in FIG. 7, it is assumed that the existence probability of the target area 36 is “0” and the existence probability of the adjacent area 37 is “0.5”. In this case, the vector 41 for the adjacent region 37 for the target region 36 is a vector having a length of 0.5 in the direction from the target region 36 to the adjacent region 37.
As shown in FIG. 8, the primary vector calculation unit 23 calculates the sum of vectors 41 for the surrounding eight regions 34 for the target region 36 as a primary vector 42 of the target region 36. That is, the primary vector calculation unit 23 calculates the primary vector 42 using Equation 1.

In Equation 1, the vector a ^→ ₀ is the existence probability of the target region 36. The vector a ^→ _ij is the existence probability of the adjacent region 37. The variable i represents the position of the region 34 in the horizontal direction, and the variable j represents the position of the region 34 in the vertical direction. Therefore, the vector 41 is (a ^→ _ij− a ^→ ₀ ). The vector b ^→ ₀ is the primary vector 42 of the target area 36.

(Step S304: Rounding process)
The primary vector calculation unit 23 changes the primary vector 42 to 0 when the length of the primary vector 42 calculated in step S302 is shorter than the primary threshold.

(Step S305: End determination processing)
The primary vector calculation unit 23 determines whether or not the primary vector 42 has been calculated for all the regions 34 selected as the determination region 35 in step S301.
The primary vector calculation unit 23 ends the process when the primary vectors 42 are calculated for all the regions 34. On the other hand, if not, the primary vector calculation unit 23 returns the process to step S302.

(Step S40: secondary vector calculation process)
The secondary vector calculation unit 24 calculates a secondary vector 43 for the first map 31 whose resolution has been reduced in step S20.

With reference to FIG. 9, the secondary vector calculation process according to the first embodiment will be described.
(Step S401: attention area extraction processing)
As illustrated in FIG. 10, the secondary vector calculation unit 24 extracts two or more adjacent regions 34 as the first region of interest 38 from the region 34 selected as the determination region 35 in step S <b> 301. In FIG. 10, a total of four regions 34, two vertically and two horizontally, are extracted as the first region of interest 38.

数２では、ベクトルｂ_ｉｊは各領域３４の１次ベクトル４２である。変数ｉは横方向の領域３４の位置を表し、変数ｊは縦方向の領域３４の位置を表す。変数ｉ，ｊの範囲は、第１注目領域３８の範囲である。ベクトルｂは２次ベクトル４３である。(Step S402: Vector calculation process)
As shown in FIG. 11, the secondary vector calculation unit 24 calculates the sum of the primary vectors 42 for each region 34 included in the first region of interest 38 extracted in step S <b> 401 as the secondary vector 43. That is, the secondary vector calculation unit 24 calculates the secondary vector 43 by combining the primary vectors 42 for each region 34 according to Equation 2.

In Equation 2, the vector b _ij is the primary vector 42 of each region 34. The variable i represents the position of the region 34 in the horizontal direction, and the variable j represents the position of the region 34 in the vertical direction. The range of the variables i and j is the range of the first attention area 38. The vector b is a secondary vector 43.

(Step S403: Rounding process)
The secondary vector calculation unit 24 changes the secondary vector 43 to 0 when the length of the secondary vector 43 calculated in step S402 is shorter than the secondary threshold.

(Step S404: end determination processing)
The secondary vector calculation unit 24 determines whether or not the length of the secondary vector 43 is zero.
If the length of the secondary vector 43 is 0, the secondary vector calculation unit 24 returns the process to step S401 to extract another first region of interest 38. On the other hand, the secondary vector calculation part 24 complete | finishes a process, when that is not right.

(Step S50: Similar region search processing)
The determination unit 25 searches for an area of the second map 32 having a high similarity with the first attention area 38 extracted in step S401.

With reference to FIG. 12, the similar area search processing according to the first embodiment will be described.
(Step S501: attention area extraction processing)
As illustrated in FIG. 13, the determination unit 25 extracts two or more adjacent areas 34 as the second attention area 39 from the second map 32 whose resolution has been reduced in step S <b> 20. The second region of interest 39 extracted here is the same size as the first region of interest 38 extracted in step S401. That is, the second region of interest 39 extracted here and the first region of interest 38 extracted in step S401 have the same number of regions 34 included in the vertical direction and the number of regions 34 included in the horizontal direction. is there.

(Step S502: First vector calculation process)
The determination unit 25 causes the primary vector calculation unit 23 and the secondary vector calculation unit 24 to calculate the secondary vector 43 for the second region of interest 39 extracted in step S501.
The calculation method of the secondary vector 43 is as described above. That is, first, the primary vector calculation unit 23 calculates the primary vector 42 of each area 34 included in the second attention area 39. That is, the primary vector calculation unit 23 calculates each region 34 as the target region 36, and calculates the sum of the vectors 41 for the eight surrounding regions 34 with respect to the target region 36 as the primary vector 42 of the target region 36. Then, the secondary vector calculation unit 24 calculates the sum of the primary vectors 42 for each area 34 included in the second attention area 39 as a secondary vector 43.

数３では、ベクトルＡ^→は第１注目領域３８についての２次ベクトル４３である。ベクトルＢ^→は第２注目領域３９についての２次ベクトル４３である。ｃｏｓ（Ａ^→，Ｂ^→）は、第１注目領域３８についての２次ベクトル４３と、第２注目領域３９についての２次ベクトル４３とのコサイン類似度である。(Step S503: First similarity calculation process)
The determination unit 25 calculates the cosine similarity between the secondary vector 43 for the first region of interest 38 calculated in step S402 and the secondary vector 43 for the second region of interest 39 calculated in step S502.
Specifically, the determination unit 25 calculates the cosine similarity between the secondary vector 43 for the first region of interest 38 and the secondary vector 43 for the second region of interest 39 using Equation 3.

In Equation 3, the vector A ^→ is a secondary vector 43 for the first region of interest 38. The vector B ^→ is a secondary vector 43 for the second region of interest 39. cos (A ^→ , B ^→ ) is a cosine similarity between the secondary vector 43 for the first region of interest 38 and the secondary vector 43 for the second region of interest 39.

(Step S504: First similarity determination process)
The determination unit 25 determines whether or not the cosine similarity calculated in step S504 is smaller than the similarity threshold.
When the cosine similarity is smaller than the similarity threshold, the determination unit 25 determines that the first attention area 38 and the second attention area 39 correspond to each other, and advances the processing to step S505. At this time, 1 is set to the variable k. On the other hand, if not, the process proceeds to step S511. At this time, 0 is set to the variable k.

(Step S505: Area shift processing)
The determination unit 25 determines, from the first map 31 whose resolution has been reduced in step S2, another first attention area 38 that is close to the first attention area 38 in the reference direction (here, the first attention area 38 ′ for convenience). Called). In addition, the determination unit 25 determines, from the second map 32 whose resolution has been reduced in step S20, another second attention area 39 (here, the second attention area for convenience) that is close to the second attention area 39 in the reference direction. 39 ′).
As shown in FIG. 14, the first region of interest 38 and the first region of interest 38 ′ may be adjacent to the first region of interest 38 ′ adjacent to the first region of interest 38. Moreover, as shown in FIG. 15, a part may overlap with 1st attention area | region 38 'adjacent to the 1st attention area 38. As shown in FIG. Further, as shown in FIG. 16, the first attention area 38 ′ adjacent to the first attention area 38 may be spaced. The same applies to the second attention area 39 ′ adjacent to the second attention area 39.
However, the positional relationship between the first region of interest 38 and the first region of interest 38 'is the same as the positional relationship between the second region of interest 39 and the second region of attention 39'. That is, if the first region of interest 38 ′ is adjacent below the first region of interest 38, the second region of interest 39 ′ is also adjacent below the second region of interest 39.

(Step S506: second vector calculation process)
The determination unit 25 causes the primary vector calculation unit 23 and the secondary vector calculation unit 24 to calculate the secondary vector 43 for the first attention area 38 and the second attention area 39 extracted in step S505.
The calculation method of the secondary vector 43 is as described above. That is, first, the primary vector calculation unit 23 calculates the primary vector 42 of each area 34 included in the first attention area 38. That is, the primary vector calculation unit 23 calculates each region 34 as the target region 36, and calculates the sum of the vectors 41 for the eight surrounding regions 34 with respect to the target region 36 as the primary vector 42 of the target region 36. Then, the secondary vector calculation unit 24 calculates the sum of the primary vectors 42 for each region 34 included in the first attention region 38 as a secondary vector 43. Similar processing is executed for the second region of interest 39, and a secondary vector 43 is calculated.

(Step S507: Second similarity calculation process)
The determination unit 25 calculates the cosine similarity between the secondary vector 43 for the first region of interest 38 calculated in step S506 and the secondary vector 43 for the second region of interest 39.
The cosine similarity calculation method is the same as that in step S503.

(Step S508: Second similarity determination process)
The determination unit 25 determines whether or not the cosine similarity calculated in step S507 is smaller than the similarity threshold.
When the cosine similarity is smaller than the similarity threshold, the determination unit 25 determines that the first attention area 38 extracted in step S505 corresponds to the second attention area 39 extracted in step S505. The process proceeds to S509. At this time, 1 is added to the variable k. On the other hand, if not, the process proceeds to step S511. At this time, 0 is set to the variable k.

(Step S509: Continuous determination process)
The determination unit 25 determines whether or not the variable k is the reference number N. In other words, the determination unit 25 determines whether or not the reference number (N) of the first attention area 38 and the second attention area 39 correspond continuously.
If the variable k is the reference number N, the determination unit 25 advances the process to step S510. On the other hand, the determination part 25 returns a process to step S505, when that is not right.

(Step S510: Matching process)
The determining unit 25 determines that the adjacent reference number of first attention areas 38 for the first map 31 and the adjacent reference number of second attention areas 39 for the second map 32 indicate the same position. And the determination part 25 obtains the conversion amount for making the 1st map 31 and the 2nd map 32 correspond from the positional relationship of the 1st attention area 38 and the 2nd attention area 39 which were determined to show the same position. .
Specifically, the conversion amount includes a movement amount that translates the map and a rotation amount that rotates the map. The amount of movement corresponds to the displacement of the positions of the first attention area 38 and the second attention area 39 that are determined to indicate the same position. The rotation amount corresponds to the angle at which the first map 31 is rotated in step S90 described later.

(Step S511: Second Area Determination Process)
The determination unit 25 determines whether or not all areas of the second map 32 have been extracted as the second attention area 39.
If the determination unit 25 has extracted all the regions, the process proceeds to step S513. On the other hand, the determination part 25 advances a process to step S512, when that is not right.

(Step S512: proximity region extraction process)
The determination unit 25 extracts another second region of interest 39 close to the second region of interest 39 from the second map 32 whose resolution has been reduced in step S20. Then, the determination unit 25 returns the process to step S502.

(Step S513: Uncertain processing)
The determination unit 25 determines that the region 34 corresponding to the first region of interest 38 selected in step S <b> 401 is not in the second map 32. That is, it is determined that the region 34 that indicates the same position as the first region of interest 38 selected in step S401 is not in the second map 32.

(Step S60: specific determination process)
In step S50, the determination unit 25 determines whether or not an area of the second map 32 having a high similarity with the first attention area 38 extracted in step S401 has been specified.
The determination part 25 complete | finishes a process, when the area | region of the 2nd map 32 with high similarity is specified. If not, the determination unit 25 advances the process to step S70.

(Step S70: first area determination process)
The determination unit 25 determines whether or not all the regions 34 included in the determination region 35 are selected as the first attention region 38 in step S401.
When all the areas 34 are not selected as the first attention area 38, the determination unit 25 returns the process to step S40 to select a new first attention area 38. On the other hand, the determination part 25 advances a process to step S80, when that is not right.

(Step S80: rotation determination process)
The determination unit 25 determines whether the first map 31 has been rotated 360 degrees.
When the first map 31 is rotated 360 degrees, the determination unit 25 determines that the first map 31 and the second map 32 do not overlap, and ends the process. On the other hand, the determination part 25 advances a process to step S90, when that is not right.

(Step S90: Map rotation processing)
The determination unit 25 rotates the first map 31 by a reference angle. And the determination part 25 returns a process to step S30, and calculates the primary vector 42 of the 1st map 31 anew.

The process of rotating the first map 31 will be described with reference to FIG.
Here, it is assumed that the first map 31 includes a frame layer 51 that defines the region 33 and a map layer 52 that indicates the existence probability. To rotate the first map 31 is to rotate only the map layer 52 without rotating the frame layer 51.
That is, the coordinates before rotation are (X0, Y0), the rotation center coordinates are (CX, CY), the coordinates after rotation are (X1, Y1), and the rotation angle is θ. Then, rotating the first map 31 is the calculation shown in Equation 4. That is, rotating the first map 31 calculates the coordinates of the area 33 before rotation corresponding to the area 33 from the coordinates of the area 33 after rotation, and rotates the existence probability of the area 33 before the rotation. It is to set as the existence probability of the subsequent area 33.

*** Effects of Embodiment 1 ***
As described above, the map processing apparatus 10 according to the first embodiment uses the difference in the existence probability that an object exists in each region 34 as the vector 41, and compares the vector 41 with the cosine similarity, thereby comparing the first map 31. And a corresponding point between the second map 32 and the second map 32 are specified.
Since the calculation is performed using the vector 41, the processing time can be shortened compared to the case of using a histogram as in Patent Document 1. Even if there is a moving object, the influence of the movement on the vector 41 is small. Therefore, even when there is a moving object, it is possible to specify the corresponding points on the first map 31 and the second map 32 with high accuracy.

*** Other configurations ***
<Modification 1>
In the first embodiment, two maps of the first map 31 and the second map 32 are acquired in step S10 of FIG. However, in step S10 of FIG. 2, three or more maps may be acquired. In this case, the map processing apparatus 10 should just perform the process after step 2 of FIG. 2 about each combination of two maps.

<Modification 2>
In the first embodiment, the existence probability of “1”, “0”, and “0.5” is set in each region 33 of the first map 31 and the second map 32. However, the present invention is not limited to this, and a finer probability may be set for each region 33.
In this case, the resolution changing unit 22 may set the highest probability among the probabilities of the region 33 included in the new region 34 as the probability of the new region 34 in step S20 of FIG.

<Modification 3>
In the first embodiment, the function of each functional component of the map processing apparatus 10 is realized by software. As a third modification, the function of each functional component of the map processing apparatus 10 may be realized by hardware. The third modification will be described with respect to differences from the first embodiment.

With reference to FIG. 18, the structure of the map processing apparatus 10 which concerns on the modification 3 is demonstrated.
When the function of each functional component is realized by hardware, the map processing apparatus 10 includes a communication interface 14 and an electronic circuit 15. The electronic circuit 15 is a dedicated electronic circuit that realizes the functions of the functional components of the map processing apparatus 10 and the functions of the memory 12 and the storage 13.

The electronic circuit 15 is assumed to be a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, a logic IC, a GA (Gate Array), an ASIC (Application Specific Integrated Circuit), or an FPGA (Field-Programmable Gate Array). Is done.
The function of each functional component may be realized by one electronic circuit 15, or the function of each functional component may be distributed to a plurality of electronic circuits 15.

<Modification 4>
As a fourth modification, some functions may be realized by hardware, and other functions may be realized by software. That is, some of the functional components of the map processing device 10 may be realized by hardware, and other functions may be realized by software.

  The processor 11, the storage device 12, and the electronic circuit 15 are referred to as a processing circuit. That is, regardless of whether the map processing apparatus 10 is configured as shown in FIG. 1 or FIG. 18, the functions of the functional components are realized by the processing circuit.

Embodiment 2. FIG.
The second embodiment is different from the first embodiment in that the first map 31 and the second map 32 are combined. In the second embodiment, this different point will be described, and the description of the same point will be omitted.

*** Explanation of configuration ***
With reference to FIG. 19, the structure of the map processing apparatus 10 which concerns on Embodiment 2 is demonstrated.
The map processing apparatus 10 is different from the map processing apparatus 10 shown in FIG. The map synthesizing unit 26 is realized by software in the same manner as other functional components. Alternatively, the map composition unit 26 may be realized by hardware.

*** Explanation of operation ***
With reference to FIG. 20, operation | movement of the map processing apparatus 10 which concerns on Embodiment 2 is demonstrated.
(Step S1: Map comparison process)
The map processing device 10 executes the processing described based on FIG. 2 and calculates the conversion amount for synthesizing the first map 31 and the second map 32.

(Step S2: Composition processing)
The map composition unit 26 composes the first map 31 and the second map 32 based on the conversion amount calculated in step S1, and generates a composite map 61.
Specifically, the map composition unit 26 converts the second map 32 based on the conversion amount. Then, the map composition unit 26 generates a composite map 61 by combining the first map 31 and the converted second map 32.
When combining the first map 31 and the converted second map 32, the map composition unit 26 adds a portion of the second map 32 that is not included in the first map 31 to the first map 31. The map composition unit 26 may use either the first map 31 or the second map 32 for the portion included in both the first map 31 and the second map 32, or the first map You may take the average value of 31 and the 2nd map 32, etc.

*** Effects of Embodiment 2 ***
As described above, the map processing apparatus 10 according to the second embodiment synthesizes the first map 31 and the second map 32. As described in the first embodiment, the map processing apparatus 10 can specify the corresponding points between the first map 31 and the second map 32 with high accuracy in a short processing time. Therefore, the map processing apparatus 10 can generate the composite map 61 in which the first map 31 and the second map 32 are synthesized with high accuracy in a short processing time.

Embodiment 3 FIG.
The third embodiment is different from the second embodiment in that driving assistance is performed based on the composite map 61. In the third embodiment, this different point will be described, and the description of the same point will be omitted.

*** Explanation of configuration ***
With reference to FIG. 21, the structure of the map processing apparatus 10 which concerns on Embodiment 3 is demonstrated.
The map processing apparatus 10 is different from the map processing apparatus 10 shown in FIG. 19 in that the driving support unit 27 is provided. The driving support unit 27 is realized by software in the same manner as other functional components. Alternatively, the driving support unit 27 may be realized by hardware.

*** Explanation of operation ***
With reference to FIG. 22, the operation of the map processing apparatus 10 according to the third embodiment will be described.
The processing from step S1 to step S2 is the same as in the second embodiment.

(Step S3: Driving support process)
The driving support unit 27 provides driving support for the moving body based on the composite map 61. Specifically, the driving support unit 27 controls the moving body based on the composite map 61 to realize automatic driving. Alternatively, the driving support unit 27 provides information on the composite map 61 to the driver of the moving body. For example, the driving support unit 27 provides the information of the synthetic map 61 to the driver of the moving body by displaying the information of the synthetic map 61 on a display device mounted on the moving body.

*** Effects of Embodiment 3 ***
As described above, the map processing device 10 according to the third embodiment performs driving support based on the composite map 61. As described in the second embodiment, the map processing apparatus 10 can generate the composite map 61 with high accuracy in a short processing time. Therefore, the map processing apparatus 10 can perform driving support with high real-time characteristics based on the accurate composite map 61.

*** Other configurations ***
<Modification 5>
In the third embodiment, the map processing apparatus 10 includes the driving support unit 27. However, the driving support unit 27 may be provided in a driving support device different from the map processing device 10. In this case, the driving support apparatus obtains the composite map 61 generated in step S2 of FIG. 22 from the map processing apparatus 10 and performs driving support.

  DESCRIPTION OF SYMBOLS 10 Map processing apparatus, 11 Processor, 12 Memory, 13 Storage, 14 Communication interface, 15 Electronic circuit, 21 Acquisition part, 22 Resolution change part, 23 Primary vector calculation part, 24 Secondary vector calculation part, 25 Determination part, 26 Map synthesis unit, 27 Driving support unit, 31 1st map, 32 2nd map, 33 region, 34 region, 35 determination region, 36 target region, 37 adjacent region, 38 1st region of interest, 39 2nd region of interest, 41 Vector, 42 Primary vector, 43 Secondary vector, 51 layers, 52 layers, 61 Composite map.

Claims (11)

For each of the first map and the second map indicating the existence probability that an object exists in each area, the existence probability for the target area and the adjacent area adjacent to the target area with at least a part of the area as the target area A first vector calculation unit that calculates a difference between the existence probability of the target region as a vector for the adjacent region with respect to the target region and a sum of vectors with respect to surrounding regions for the target region as a primary vector of the target region When,
For each of the first map and the second map, two or more regions are used as a region of interest, and the sum of the primary vectors of the regions included in the region of interest is calculated as a secondary vector of the region of interest 2 A next vector calculation unit;
Comparing the secondary vector calculated for the region of interest for the first map with the secondary vector calculated for the region of interest for the second map; A map processing apparatus comprising: a determination unit that determines whether or not the region of interest for and the region of interest for the second map correspond to each other. The determination unit approaches the first map when the adjacent reference region of interest for the first map corresponds to the adjacent reference region of interest for the second map. The map processing device according to claim 1, wherein the reference region of interest and the reference region of interest adjacent to the second map are determined to indicate the same position. The determination unit determines whether the reference region of interest close to the reference direction for the first map corresponds to the reference region of interest for the second map adjacent to the first map. The map processing device according to claim 2, wherein the map is determined while rotating the map by a reference angle. The determination unit includes a cosine similarity between the secondary vector calculated for the region of interest for the first map and the secondary vector calculated for the region of interest for the second map. The map processing device according to any one of claims 1 to 3, wherein the map processing devices are compared by calculating. The primary vector calculation unit selects, for each of the first map and the second map, from the second region from the outside to the region of the number of objects inward as the target region. The map processing device according to any one of the above. The primary vector calculation unit sets the primary vector of the target area to 0 when the sum of vectors for the surrounding area of the target area is smaller than a first threshold;
The secondary vector calculation unit sets the secondary vector of the region of interest to 0 when the sum of the primary vectors of the regions included in the region of interest is smaller than a second threshold;
The map processing apparatus according to any one of claims 1 to 5, wherein the determination unit compares the attention areas whose secondary vectors are not zero. The map processing device further includes:
For the first map and the second map, a resolution changing unit for reducing the resolution of the first map and the second map by making a plurality of areas into one area,
7. The primary vector calculation unit according to claim 1, wherein the primary vector calculation unit calculates the primary vector for the first map and the second map that have been reduced in resolution by the resolution changing unit. Map processing device. The map processing apparatus according to claim 7, wherein the resolution changing unit sets the highest existence probability among the existence probabilities for each of the plurality of areas as the existence probability for the one area. The map processing device further includes:
A map combining unit that combines the first map and the second map to generate a combined map based on the attention area determined to be supported by the determination unit;
A driving support unit that controls a moving body or provides information to a driver of the moving body based on the synthetic map generated by the map combining unit. The map processing device described in 1. For each of the first map and the second map indicating the existence probability that an object exists in each area, the existence probability for the target area is adjacent to the target area, and at least a part of the area is adjacent to the target area. The difference between the existence probability for the adjacent region is a vector for the adjacent region for the target region, and the sum of the vectors for the surrounding regions for the target region is calculated as a primary vector of the target region,
For each of the first map and the second map, the computer uses two or more regions as a region of interest, and a sum of the primary vectors of the regions included in the region of interest as a secondary vector of the region of interest Calculate
The computer compares the secondary vector calculated for the region of interest for the first map with the secondary vector calculated for the region of interest for the second map, and A map processing method for determining whether or not the attention area for the first map corresponds to the attention area for the second map. For each of the first map and the second map indicating the existence probability that an object exists in each area, the existence probability for the target area and the adjacent area adjacent to the target area with at least a part of the area as the target area A first vector calculation process for calculating a difference between the existence probability of the target region as a vector for the adjacent region with respect to the target region and a sum of vectors with respect to surrounding regions for the target region as a primary vector of the target region When,
For each of the first map and the second map, two or more regions are used as a region of interest, and the sum of the primary vectors of the regions included in the region of interest is calculated as a secondary vector of the region of interest 2 Next-vector calculation processing,
Comparing the secondary vector calculated for the region of interest for the first map with the secondary vector calculated for the region of interest for the second map; The map processing program which makes a computer perform the determination process which determines whether the said attention area | region about 2 and the said attention area | region about the said 2nd map respond | correspond.

Images